Hypertrophy Past and Present (72) Why you’re probably reading hypertrophy studies wrong

0:01 Welcome back. Thank you for joining us for another episode of Hypertrophy Past and present. Chris, how are you doing today? >> Yeah, I'm doing well, thanks Jake. >> I've searched long and hard to try to find a routine to honor our guests who we had a couple weeks ago now, Saurin. And I think I found the routine. So, a lot of we had some feedback on actually we had some feedback saying, "Hey, those exercises are really interesting. I'm going to try them in my gym." and people actually wanting to to have visual aid on how to do them and and getting really experimental. And I had some people sending me videos of exercises they were now coming up with. So, it definitely has inspired a lot of people. And again, if you've listened to that episode and you're like, I I needed help visualizing those exercises, go follow Saurin's page because he does post a lot of these videos. But it's not new. It's not a new development for people to experiment with exercises and different exercises and and you know unconventional exercises. And I actually posted a routine yesterday on social media. And no, it wasn't a routine. I I posted like some some commonalities you see in silver airlifting. And one thing I'd mentioned was that most of them were doing a aaa. And I got this comment back and someone goes, "Not disagreeing with you, but I think they did that just because there wasn't much exercise variety. They only had dumbbells and barbells and they they perhaps just couldn't think of more exercises to do."

1:24 And like it was an innocent question or statement and you know, whatever. But I immediately laughed because there's books that are written in the 30s, 40s, 50s that have hundreds and hundreds of different exercises listed in them. And often I'll look at them and I'll be like, "What the hell is that exercise?" Or I'll find diagrams, you be like, "Man, I would have never in a million years have even thought of doing an exercise like that." So I think creativity and imagination was definitely not lacking for some of these guys. So that is our our my inspiration for today's routine.

1:55 Now the routine we've got it was it is a 1950s routine. Okay. So this was I think it was 1956 or thereabouts. he came from magazine and it was written by Bart or Bart and Hovath. So I don't think he ever won competitions but he did win like you know there was like subcategories in the bodybuilding comp. I think he won like most developed man or some body part was most developed like he was definitely a competitive athlete or competitive bodybuilder and he wrote this series of of different you know magazine articles and one of them was on I think he called them the what did he call them the the the unusuals I think he said and so these were exercises where he was like look if you're if you're getting stale or you've hit a plateau it might be worth adding in some of these unconventional type exercises exercises. Now, I don't think the idea here was to do these permanently and and for this to replace your routine forever. I think it was more again being a bit imaginative trying to find and he would talk about like you will introduce some of these exercises and it might expose maybe you know a weak area for you maybe you know maybe there might be an exercise there that you then maintain right so that's kind of the idea behind this routine now it is like I've warned you guys this is an unusual routine literally in the name it was called the unusual so brace yourself it's full body. Again, it was done as a AAA, which was pretty much a norm. And we begin with probably one of the one of the the least odd exercises, and it's still quite odd. What was called a dumbbell roll out. Now, this was actually done I've seen this in a few routines at this time. And the way dumbbell roll out was done would be imagine like imagine you do a dumbbell fly, but you're facing the floor. And so, you're actually starting essentially in a push-up position with dumbbells in your hands. and the dumbbells are rolling out and your your arms are just going directly out to the sides and then you're trying to kind of do the flying motion, bring yourself back up again. So ultimately, I guess you could think of that sort of as like a almost like a body weight fly, if you will.

4:12 The next exercise that was used was a barbell extension done behind the back. So, this actually I think came from Sig Klein. I think we've talked about this exercise before. I don't know if if Sig came up with the idea, but it's the earliest I've seen of it. So, essentially it was literally just a barbell behind the back and then you're you're fully extending behind your back and it' be more like a triceps type exercise. Then we have a suited suited a seated goose neck dumbbell curl which again is actually an exercise that was used in the Bronze Zero. So, he's kind of borrowing some of these very old exercises. And I think we actually talked about this in our perhaps in our first ever episode cuz we were talking about the Milo barbell company. And they used to use this this exercise a fair bit. So, it's basically just reverse curl, but you you kind of roll your your wrist forwards, which actually feels quite comfortable when you do it like that. and it sort of makes your grip less of a limitation the exercise. So, basically just a a reverse curl.

5:16 then we've got a one arm get up, which I do people still do that today? I feel like that's like a, you know, in the in the quote functional world, I feel like that's an exercise people do where they'll hold a barbell above them and they're lying flat on the ground. They get up holding this barbell. I've never attempted to do it. I I'm not going to talk on that as an exercise, but that was what he suggested next.

5:38 Then we have a kneeling clean. Now, I thought this was quite interesting. So, and I've not tried to do a kneeling clean. Chris, I don't know if you've tried to do a kneeling clean, but I imagine we're going to get I mean, you'll tell us anyway, but I imagine this would be an okay anterior del type exercise and it's exactly how it sounds like you're you're kneeling and I I don't think he used a lot of momentum. I I did see an image of this exercise being done. It didn't look like it was this full sort of, you know, dynamic Olympic lift type thing like it it looked more controlled. So, I don't know how it differs from a front raise. I'm not entirely sure. But that was the next exercise. Then we have a onelegged squat. And if that's not funky enough for you, you had to hold a barbell overhead.

6:28 And this is maybe where it descends a little bit into a bit of madness. We then have a twofinger deadlift, which I didn't know this, but this was kind of like a whole thing. Like there were people essentially like competing in like two-finger deadlifting. Like there's magazine articles written about people two-finger deadlift deadlifting using immense loads, massive loads, just holding them with two fingers. And he I mean actually this is what made me think of Saurin because he did talk about actually strengthening the fingers. This was part of the purpose of doing the the two-finger deadlift. So sorry and there you go. Finger building has been part of bodybuilding for almost 100 years. And then finally we concluded with maybe the most sane one of the more sane exercises here which was a barbell sit up. And we don't talk about abs a lot and a lot of people ask us to talk about abs. And here you go. This is perhaps not a bad ab exercise. Literally it's a sit up and he was just holding a barbell at arms length. What do you think of our unusuals?

7:29 >> Yeah. Well, they're definitely unusual. yeah. I mean, I guess with a couple of exceptions, the the unusualness is largely a lack of stability. >> Yeah. For for most of them, half of them, maybe a bit more. Yeah. somewhere between a half and two/3 of the exercises it's the unusualness is is it's not a very stable exercise. So it becomes extremely coordination dependent activity with probably relatively low relatively low agonist motorunit recruitment levels compared to a similar more stable variation.

8:20 So yeah, I mean it's it's it's not really a bodybuilding workout. >> You know, on that point, the point of the stability and coordination demands of these exercises, I forgot to say how many sets he would do, and he wasn't overly specific with the sets. But what he said is you would do a set kind of essentially like you'd kind of practice the movement. You do the set, you'd add a little bit more weight, you do another set, and you just did that till you got to your max effort or max load that you were able to do. And I think that's an interesting statement in and of itself because that's not super common at this at this time. Maybe it's becoming a little bit more common, but obviously typically people would either do single sets or they would do multi sets with the same load. There wasn't this idea of ramping up load very much at this time.

9:11 So I wonder if that is just a product of hey these are highly coordination based skills and you actually need to practice them. >> It kind of fits more with a bronze era kind of mentality of practicing specific tricks >> rather than and no the two-finger deadlift is right there in that in that category. I mean it's literally just sort of >> I can do this and you can't do this. >> It's a circus trick. That's what it is.

9:37 Yeah. >> Yeah. So, it's not a bodybuilding workout. It's it's a collection of of exercises that you can do so you can impress people who can't do them. >> Yeah. Is there anything there that you think? So, like the dumbbell roll out, what do you what do you think of that? Cuz it's not like that's necessarily super unstable, right? Like you're in contact with the ground. I guess the dumbbells are adding a degree of instability. So the thing is I was trying to work through in my mind why I thought that would be quite unpleasant on the eccentric phase and I think the reason is because on the concentric phase you've got a natural point where it's going to stop which is when it hits the dumbbell of the other hand.

10:26 >> So you actually as you bring the dumbbells together you're going to stop in the middle and it's the the energy of the two sides is going to collide and you're going to stop. That's not applicable when you go in the opposite direction. >> No. >> So, you're I think you're naturally going to want to be much more controlled on the eentric than you are in the constant. It's just going to it's going to feel like it's running away with you.

10:47 I think >> like ab wheel rollouts really. >> Yeah. >> So, I think I would expect people to get quite sore doing that. >> Yeah. Yeah, that's true. Yeah. It's probably going to be more of an eccentric type movement, isn't it? And Yeah. Yeah. Yeah, I guess it is like an ab wheel roll out for the pecs. The only exercise I'm really curious about if people have tried it is the the is the shoulder extension behind the back.

11:19 >> if people have tried that and they've made it work, I'd be interested to know. the moment arms don't really work well for me. but yeah, if people have tried that and they've made it work, I'd be I'd be curious to see or hear how it's how it feels. But >> it's interestingly, it's an exercise that has was used a little bit. >> I don't see it at all now, though. >> No. again, like I mentioned, C.

11:46 Klein had done a version of it. And when I do see people typically do it, like when I say see people, I mean see magazine snippets and stuff from 100 years ago. Typically what they're doing is they're they're leaning forwards. >> Exactly. And that while they're doing it, they're fixing the moment arm problem that I was referencing because it's basically you're going to have the same problem you have with lateral raises, which is that you you start doing the exercise and before you know where you are, it's, you know, it's impossible to carry on.

12:14 >> Yeah. >> You know, so it it just the moment arm change is huge. but when you start, it's easy to get going. And that's the same issue you're going to have with with that. It's just obviously a different plane of motion. But yeah, I mean, if you were to lean forward or, you know, set it up in a way that allowed you to have an inclined torso or an inclined body, then you know, it might be >> it would be similar to a kickback just without bending the elbow >> and more stable. Yeah. I mean, I wonder whether you could do it like on a 45 degree back extension or something similar where you >> possibly.

12:51 Yeah, I wonder. I've never tried it on a back extension. I've done it just in a cable station and like honestly I didn't really mind it, but it was a type of exercise where if you do low reps, it's just like a heavy load >> and you just been pulled all over and Yeah, it's not happening. >> Okay. Interesting. any other comments on any of these exercises? Barbell, sit up. Anything you want to say about that? It's not really unusual, is it?

13:18 >> I mean, it's a weighted situp. I mean, >> yeah, there you go. >> Yeah. Kneeling clean. Do you have any thoughts on >> I'm not even going to go there. >> He talked about using tremendous loads of that. I think he talked about about 100 kilos, which it's interesting cuz you're taking out the body English. Like there's no real momentum if you're literally kneeling. So, I don't know. It feels like one of those exercises where I see it, I'm like, I might have to give it a go. There might be something to this.

13:49 >> I know. no. >> No, don't don't don't do things like that. >> Okay, I'll report back on my kneeling clean in a future episode. >> Just get hurt. It's just not clever. >> Okay, moving on. I think that's it for our our unusuals. It was an unusual plan. >> The short the shortest conversation we've had about a a program ever on this podcast. Well, I thought I needed to give you a counter perspective when you say how much more sane the silver hero plans were to the golden era. Well, now you could remember this one.

14:24 >> Okay. Temper temper the excitement for the silver hero. >> Okay. I'm going to let you introduce today's topic even though I'm the one who request. Yeah, >> that's very kind of you. Okay. So, yes, at Jake's request, what we're going to do is talk through some measurements, that are commonly used in the strength training literature and, just give a little bit of perspective on them. So, I'm emphasizing perspective because you know, especially in this day and age with AI, if you want to know what a measurement is actually made up of and how it works, you can just ask your favorite AI agent and it'll tell you pretty much exactly how it works.

15:11 But if you want some perspective on it, then that's what I want to do in today's podcast. So just try and give an insight into the way I see it and you know how that then allows me to kind of connect areas of research with each other >> and just for my two cents as to why I push you to do this topic. So I'm part of a few different groups and communities and and stuff of people who I guess largely are either coaches themselves or are that kind of so you know when someone's so personally interested in a topic that they become like a little bit obsessive. So there's a lot of people in that kind of category.

15:50 >> I have no idea what you're >> You have no idea what that's like? No. I didn't think you would. And in these groups and these these these sort of group chats and stuff like that, a lot of the time people will be pinging back and forth research and and studies and there's often conversations and questions about like well you know what is what does it mean for for this to change or they use this in the methodology what does that mean or you know how accurate is this as a measurement and I love seeing people go to the literature and go to research and trying to understand these things but then I feel like a lot of people are getting getting caught up in I don't actually know what that thing that's being measured tells me. And so that was the motivation. So if you're someone who does do that and you go to literature and you're like, you know, I want to learn for myself. I don't want to just take someone's word for it, but now I'm stuck looking at, oh, they use EMG for this and I don't know, is that even accurate? I don't know. Well, that's kind of the idea behind what we're going to talk about today.

16:46 >> Awesome. and so we've got a list of areas that we're going to work our way through. before we jump in, I'm going to add my two cents, which is that the single most the single biggest problem that people are going to face when they start looking at literature in the context of a group conversation. So in the context you described where people are pinging papers backwards and forwards and discussing it and arguing about what it means and you know what it can be used to defend is the wiziarti principle of psychology. So whizarti stands for what you see is all there is. And what happens when people are arguing about a single paper is that they exclude everything else that exists in the universe and the only thing that they then focus on is this exact paper.

17:44 and the problem with that is it ends up being a an argument that doesn't get anybody anywhere. It it doesn't even if the argument is civil, it still doesn't get you anywhere because at the end of the day, all you've argued is what the paper can support, what the paper can't support. And we live in a world where there's more than one paper. So it doesn't actually get you anywhere useful. So what you've got to do all the time every time you're in those scenarios where you've got a a discussion about a specific paper is try and break the whizy spell, try and take a step back and go, what else exists? What what else exists? And that's basically what I'm doing all the time is going what else exists? to the point where I'll look at neighboring areas of literature. I gave examples in the past how I look at sumagenesis literature from you know limb immobilization studies or you know limb distraction studies or whatever else I can find that will give insight into other scenarios where exactly the same mechanisms are operating. and I think that's really the most important thing I would say up front is try and break the wiziarti spell. Try and get yourself out of the headsp space that the only thing that exists is the kind of data set in front of you. remember that there's other stuff that exists. And that doesn't mean that you kind of turn the conversation into something that is now arguing about a topic rather than about a study, but just have some priors. So, so you're not going into a study and going, well, because the study, you know, didn't control for this tiny thing or because the study used this particular measurement method, which is not as good as another particular measurement method, I'm now going to throw that study out the window. If that study actually has results that align perfectly with a whole set of really solid prior that you have, then I would be much more relaxed about those small areas of kind of weakness than I would be if the study was flying in the opposite direction of those priors. And I think that's something that just gets completely ignored in all of these conversations is people just arrive and go, "Oh, no, no, you're not allowed to have prior," which is a very evidence-based kind of tendency rather than a science-based tendency.

19:51 like as a science-based kind of practitioner, you should have pretty strong prior. You should actually have ideas about how things work that you've tested in the past and you're confident about. If you don't, then maybe you shouldn't be in WhatsApp groups arguing about the interpretation of specific studies, but that's a separate point. >> And so when you say prize and you're saying like you would have expectations as to what a result might be based on how you think the physiology works.

20:17 >> Exactly. Well, not just the physiology, but you know, the evidence base as well, like the the the training literature or the specific previous studies that you've looked at in that same area. I think if you're if you are arguing about a study and you don't have any priors, then you probably shouldn't be arguing about a study. >> Yeah. Because literally you would be treating that study as like that is the that is the only study that exists.

20:41 >> And as I say, that's actually what tends to be the default option because wizardi is a human tendency. We tend to just go the only thing that exists to the thing in front of us. And breaking that tendency is the first step to actually starting to interpret something in context. >> Cool. Okay. So cool. >> so with that behind us, I think we've got a list. We're going to start with the kind of headline concept which is hypertrophy. So I already did this kind of disclaimer I think maybe on the AI podcast, but I'm going to do it again. Hypertrophy suffers from a terminology problem in the sense that the strict definition of hypertrophy is an increase in muscle fiber diameter, but everybody uses it to refer to an increase in muscle size or mass or volume or whatever. So, there's a difference between the way that the term is technically defined and the way that the term is practically used even in the research literature. So, you have people referring researchers referring to ibottophy and what they're actually referring to is an increase in whole muscle size. technically that is not a valid statement but everybody does it so nobody really argues. so technically an increase in muscle size could be caused by a diameter increase of the fibers. It could be caused by an increase in number of fibers which is hyperplasia. Probably doesn't happen very much in humans for reasons that I've probably talked about before. sarogenesis which definitely does happen in humans. again probably doesn't occupy as much of a muscle size change as hypertrophy does though and of course an increase in intramuscular water which we'll come on to later on. So all of those things could increase muscle size and if somebody refers to increase in muscle size as hypertrophy they're technically not correct because hypertrophy is only one of those things. In an ideal circumstance it would occupy the majority of the increase in muscle size change but it probably doesn't all the time. So, in terms of measurement methods that you're going to find for whole muscle size, you're going to find basically the most common one at the moment you're going to come across is going to be an ultrasound based measurement, probably muscle thickness.

22:47 You can measure cross-sectional area with ultrasound as well, but thickness one tends to be the most common. obviously with thickness, you're literally just kind of measuring what is approximately the diameter of the of the muscle tissue. if you're assuming that it's cylindrical. and then obviously cross-sectional area, you're actually getting a genuine cross-sectional area. >> so okay, so if someone pulls up a study, they look at like here's a standard, you know, 8week hypertrophy study. One group did two times a week, one group did one time a week, and then they'll look at the the measurements and and firstly those studies are not going to be doing these measurements on every single muscle, are they? So the the the study may involve a full body program, but they're typically going to measure one or two sides.

23:33 >> So that's that's a good that's a good point to bring up at this particular moment because generally speaking, muscle thickness tends not to be as specific to a muscle as cross-sectional area is. So generally speaking, if you're measuring muscle thickness changes, like for example, one of the very very common things to do in a lot of strength training studies is to measure the quadriceps muscle thickness there will be it will just be referring to anterior thigh. So you won't actually know specifically. Now you can kind of work it out if your anatomy is solid and you can kind of look at the description and the description is good enough and you can kind of figure out roughly what muscle they're they're kind of positioning the scanner over. But it might be a combination. I mean often it tends to kind of blend together a couple of the quadriceps muscles whereas the cross-sectional area measurement will generally give you a specific muscle or you know kind of the whole group all together. So that is really really important if you're for example something we talked about before the podcast. If you're measuring something that is in a multi- joint activity like a squat and you've got rec fem involved in your measurement then you are kind of putting yourself at a disadvantage or you're going to get skewed results because the rectum is not going to do anything. So you know if you're measure if you're looking for a change in the quadriceps and you've got fem included in that and you haven't got a knee extension in your in your strength training protocol you find that you don't get quite the results you're expecting to get. So that's an important thing to bear in mind in that scenario >> which sounds super basic but it is it is in fact very important to check. I've seen multiple >> researchers do not check it and there's a lot of researchers out there who do not take that into a consideration. There's a very famous famous in the sense of you know kind of people who know about this stuff.

25:16 But there's a there's a relatively well-known study where they literally had two different training groups. one had knee extensions in and one didn't. And they measured kind of quadricep practice for Morris as their as their quadriceps outcome which basically meant the only group that got any change was the group that trained with knee extensions. And they thought that that was a feature of the program design rather than a feature of the exercise selection because they didn't know for whatever reason. I mean like this blows my mind, but for whatever reason they did not know. And this really goes back to a criticism that I just keep throwing out there, which is that when researchers don't read outside their area, they miss stuff.

25:54 >> That's what happens. So, if you focus if you if you're like a physiologist and you ignore biomechanics and then you you're going to end up making a mistake like that, which is where you end up having two different training programs that you're interested in comparing. One of them was like a pre-exhaustion program and one was a a standard strength training program. And for the pre-exhaustion, they threw a knee extension in there. and they thought that they'd have no difference as a result as a result of changing the exercise selection. But a somebody who understands biomechanics knows there's an enormous difference there. Like that's what happens if you don't actually do the reading outside of your immediate area. If you literally just do what was required for your PhD or whatever and that's the only studies you ever read, you don't go and read the psychotinist literature or you don't go and read the you know kind of biomechanics because you're focusing on something very specific, you are going to make horrendous errors like that that are just going to >> stupidly embarrassing when you find out about them.

26:45 >> And some of these like these kind of things can be quite common. You know, I've seen studies where they might be measuring the biceps, but then the program has no direct biceps work and so they're changing, you know, they might be adding more sets of compound lifts or whatever, but it's like, well, how accurate is measuring the biceps here or the elbow flexes going to be for this program? Or likewise, I've seen somewhere there might be two muscles that are being measured. Maybe let's say, you know, quads and and whatever, triceps. And then they might be looking at frequency. And then they might go from, you know, one workout to two workouts a week, but there might be one quads exercise, but then there might be like four different exercises for the triceps. And then they might conclude that actually the triceps, >> the volume. Yeah. Exactly.

27:26 >> But then the conclusion will be wrong. And they might say, well, look, this muscle benefited from high frequency, but this one didn't. It's like yeah maybe cuz you did like four times more sets. So this becomes really important to look at what is the workout being done in these studies and then what muscle which muscles are being measured. >> Yeah. Exactly. And you know that all came from the difference between muscle thickness and and muscle cross-sectional area measurements both being measured by ultrasound. Now you can also add on MRI.

27:55 You can have CT scans and they'll give you cross-sectional areas. You can also get volume measurements generally that they're they're constructed as a series of slices of cross-sectional areas that are all summed together to produce a volume measurement. Volume measurements are considered to be the highest kind of quality because what they'll do is they'll make sure that if you've got any regional variation like for example you can get distal growth after some types of like eccentrics or stretch position exercises you get more kind of proximal or middle growth after concentrics or you know kind of whatever else that doesn't have a massive tension component. you know you're going to get variances across the length of the muscle. So volume will make sure that you don't miss that. So they're generally regarded as the highest quality. does that mean that you want to kind of just then only go and look at studies that have measured muscle volume changes? Well, no, not really because you're going to make end up only having a very tiny number of studies that you can look at. So honestly, I think just kind of bear in mind that there is a kind of spectrum of possibilities that you can look at in terms of whole muscle size measurements.

29:02 I mean, you can even if you really want to, you can kind of go all the way and say, "Well, we're going to look at lean mass changes." I mean, I really kind of draw the line at that and I don't really like though looking at those studies very much. I don't think they tell you very much about muscle. >> Typically, this would be DEXA. >> Yeah. So, you can get studies that will give you a lean mass measurement and they'll give you a leg lean mass change.

29:23 Honestly, I've never found any of those studies to be particularly helpful. that's just been my perspective. And what about muscle biopsies? >> So what you can now do is you you've kind of we've we've talked about whole muscle size changes and we've talked about the main kind of risks associated with with those. you can actually then you can do biopsies and you can take sort of measurements of single muscle fiber diameter changes or cross-sectional area changes.

29:52 Now the on the one hand on the one hand the advantage of doing that is that you've got you've got closest to what you're really interested in which is an increase in muscle fiber cross-section area or di diameter. So you've not used a proxy you've gone straight to the thing that you're most interested in. The biggest downside with that is that they are just horrible measurements to take and you're not going to find a very very solid answer. In fact, this is why most of the time when researchers actually take measurements of single fiber cross-sectional area, they don't generally find any changes because the the variability is just so gigantic.

30:35 the actual abil our ability to take those measurements is so low compared with a you know an MRI taking cross-sectional area measurement that you generally don't see a significant change and it's not because there isn't one it's because we're just really not very good at taking those measurements. >> Okay. So, so if I guess what I'm saying there is if you have a study and you don't have an increase in single fiber cross-sectional area, don't go kind of thinking that that is a really clear indication that hypertrophy did not happen.

31:09 It may well have happened. It's just >> and if they're comparing slow twitch and fast fast twitch fibers, would you like that change at I wouldn't really read too much into those measurement methods right now that they're really really not very accurate. so you know generally the only measurements used then and the study for example like you're saying you come into this study you're expecting based on everything else that this is going to be the result and then this study has found the opposing result and they've just used biopsies.

31:44 >> >> I'd be a little bit kind of discounting that. I wouldn't say that it's not telling me anything, but I'd just be a little bit discounting it based on the unreliability of those measurements. And that's that's not a I'm very forthright with my criticisms of hypert researchers. This is not a criticism of hypert researchers. It's a really really difficult measurement. >> And then muscle circumference, you would just put that in the same as lean mass.

32:10 >> Yeah. Yeah. Absolutely. probably same kind of category. It doesn't really tell you very much, I don't think. I don't think it's a very very useful one. I mean, we've kind of skipped over a little bit with cross-sectional area measurements. We skipped over the difference between anatomical versus physiological. that's not going to affect people most of the time. there's probably for every one physiological cross-section area measurement, there's probably 99 anatomical cross-section area measurements. So you're almost never going to come across it. It's only really done in biomechanical scenarios where people really want to know what the actual kind of capability of the muscle force is in any specific scenario because it gives you a cross-section area measurement that aligns with the actual direction of the fibers rather than just with the slice perpendicular to the bone that you're resting the scanner on.

33:05 So, of these different measurement types, is there a gold standard where you know we would look at that like this? >> Yeah. MRI volume. MRI volume is going to be the gold standard. Yeah, pretty much. It's going to be a collection of cross-sectional area measurements taken with MRI is going to give you some that together going to give you a volume and that's going to be the best way to be confident that you've captured all the information about muscle because the problem is if you take cross-section areas, well, where do you take it? You take it in the middle maybe probably because that's where the biggest changes tend to occur. What happens if one of your protocols was particularly good at increasing distal region and you missed that? So that's why volume is good.

33:42 >> Yeah. Like that elbow flexor study where >> Yeah. That's why I criticize the elbow the the cross-sectional area measurements that they take in the in the elbow flexor studies because they measure them in places where it's going to capture one muscle and not another muscle. And then you do a study program. >> Yeah. They just range of motion to train the other muscle. It's just like textbook errors that people make if they don't understand biomechanics when they're doing a hypertrophy study. This is why I told people that it helps to read outside of your area.

34:13 >> I mean, the reality is, and we're not going to talk about biomechanics in this episode, but ultimately that is something that you need to know when you're then looking at actually what workout routines have been used in these studies because that, as we mentioned, I guess earlier, like that's very influential to what results one would expect to see. And it's not uncommon that exercises are being used, I would say, with the intention of it training a muscle, which biomechanically we wouldn't expect it to actually do that.

34:42 >> Sure. Exactly. I mean, the range of motion ones are the classic because they just don't take into account the fact that different muscles work at different muscle different joint angles. I mean, that's just kind of the the single I mean, I went through that on the AI podcast. It's the single biggest area of error or source of error in those studies that that there is. but yeah, so yeah, I think that's probably everything on the hypertroy measurements side.

35:08 >> what about like swelling? >> Fasical length. Would you put that into this category? So obviously there's contentious obviously it's a contentious area but I think it's fairly straightforward that fascinis I think that if you've got a fasc increase in a human study that's telling you that you've got sarcimein series being added it's going to increase muscle volume it might increase muscle cross-sectional area depending on how your measurements being taken.

35:44 Mhm. Okay. Depending on how in what way >> in the sense that what location in the muscle you're taking it out. So generally speaking >> it it just comes down to the geometry. So it's not predictable where it's going to make the difference. But sometimes you might not see a difference and sometimes you might. In the same way that you if you have a regional muscle growth, you might not detect it with cross-sectional area. That's the problem with cross-sectional area is where the slice is is what data you collect. And if if your kind of vasal lengths haven't created an increase in cross-sectional area at that particular point in muscle, then you won't detect it. But if you're measuring lengths, then you will. I mean like >> So how how are they measuring a facasicle length?

36:32 >> Use the ultrasound. Yeah. So you want to be a little bit aware. I'm not saying careful, but you want to be a little bit aware of the different types of ultrasound that are available. If you can get extended field of view, that's a higher confidence that what you're detecting is definitely correct. Just skim read them. I'm assuming that people are not reading papers word for word. I mean like like just making that assumption. Maybe that's unfair, but like just skim the method section. And if it says that they used a calculation for the fascal length, then what they're doing is that they're actually not measuring end to end with the fas.

37:10 They're measuring a section of the fasculating with like a geometrical term where they think the fasicle is going to end. Now, that's not obviously as accurate as having an extended field of view where you get both ends of the fasicle. That's going to be particularly problematic in the quadriceps, not going to be so problematic in smaller muscles. Okay. Okay. Anything else with hypertrophy or that's probably about it? >> The swelling aspect is probably worth mentioning because that kind of confounds that compounds the the the issue with with hypertrophy. So basically at the moment >> so on swelling would you maybe you're going to answer that would that affect all of those measurements you just talked about or which ones of those could swelling affect? So, so okay physiological background to swelling basically swelling is primarily intracellular.

38:06 It's internal to a muscle fiber. can get extracellular kind of fluid accumulation. And many studies assume that that's again this is one of those areas where it helps to read outside of your area because if you're a hypertrophy researcher and you just make the assumption that swelling is something that happens outside of a musifier then you you know you're going to run into difficulty because the fatigue literature makes it relatively clear that it's intracellular most of it anyway. so you are looking at a single fiber increase in cross-sectional area as a result of that that swelling phenomenon. so yeah it's going to affect everything literally everything. there's not I mean you could you could argue that it doesn't affect fascical lengths but I think it would change the geometry of the muscle and it probably change your resting sum lengths. So it probably is going to affect lengths as well. So yeah, I mean I think ultimately this is why I say that the horn study that measured that basically estimated intram intramuscular both extracellular and intracellular water using biological impedance. I think that's probably the only study that's ever really tried to control forward content changes as a result of a strength training program because ultimately if you've done a pre and post muscle size measurement then your pre-measurement is not going to have any swelling effects and your post measurement is. Now they will go into sort of detail potentially depending on the study about how many days after the last workout they left before they did the kind of measurement because the idea is that if you leave it like 2 or 3 days then you shouldn't have very much muscle swelling. Again, that's ignorant because they're not looking at how fatigue literature actually describes the path of muscle swelling post-workout. It actually goes up immediately post-workout, then goes down and it actually is lowest in your kind of several days post-workout, it's probably lowest around about the 24-hour period and then it goes back up again. so if you're kind of picking a day, it's like 48 72 hours post last workout, you might actually be accidentally picking the maximum amount of muscle swelling that you could have literally apart from post final workout, you know, 30 minutes or whatever. So you have to be careful about muscle swelling because as I say until we get to a point where hypertrophy researchers take this seriously most of our hypertrophy data is probably contaminated to varying degrees and degree is the kind of area of contention but to varying degrees by some muscle swelling. And really what we need is to deduct both extracellular and intracellular water using kind of estimates like horn used in the in the extreme boring study that that was done.

40:46 >> Do we know how long swelling can persist for? Most of the studies I've seen they typically stop measuring it at what 72 hours. >> Well, it's I mean like numerically it's still there at 72 hours after a high volume strength training workout in strength trained subjects. So we we see all these people kind of making noise about say oh well you know it goes away. Well okay so why do we have strength trained subjects studies in strength training subjects showing that there is still swelling numerically of like somewhere between 5 and 10%.

41:17 >> you know 3 days after a strength training workout that involves moderate kind of volumes at least by modern standards. you know why is that happening then you know and significance isn't always reached but then you look at the confidence intervals you can see why but there's never anything below zero it's like >> everything just sits above above zero it's all and and the problem is that strength training subjects don't increase muscle size by very much you see maybe 5 to 10% increases well that's could all be muscle swelling you know so muscle swelling is a huge problem for the hypertrophy literature. And at the moment, what I see is hypertrophies are just sticking their fingers in their ears and pretending it's not a problem.

42:05 >> I know this is not the point of this podcast, but what would be one what would be the way to avoid that as an issue? I know you mentioned obviously what what Cody did and we can talk about by impedance and and measuring that in a second, but how long like if if 72 hours may not be enough for that swelling to go back to baseline, we're still potentially seeing over 5% increases like what 5 days, 7 days like when would we expect that there's not going to be any swelling affected?

42:30 >> Well, I think I mean there's a couple of ways that you could I mean like I'm not the most creative at study design. I mean, I don't think I'm necessarily going to be the person who can kind of design the perfect study that would do that. I mean, like, you could you could do a muscle swelling measurement after the first workout to get an idea of what the swelling is on an individual basis for each person.

43:01 You could >> create yourself a baseline like that. you could could do the same thing at the end of the study pre and post final final workout, see what happens. But the problem with that is that it ignores the accumulation of of of water that happens when you create accumulator fatigue. U because ultimately the water that stays inside the muscle over time past the end of a workout, it's not the hydrogen ions that are doing that, which is what happens during a workout. It's the inflammation that's doing that.

43:41 >> So you know, I mean, I think it's a genuine problem and I don't think that there's a there's an obvious answer to it. you know yeah I'm not >> you like could you just take the measurements 5 days later or 7 days later and yes it'll be some atrophy that occurs I mean yeah I mean you could go down that route but again it's going to depend on just how much it's going to depend on just how much accumulated fatigue you've created.

44:13 >> Yes. Because again, this is the this is the issue that we have when people say, "Oh, no, no, no, no. our subjects are getting used to this >> and therefore it becomes less of a problem." And I'm like, "You're not paying any attention to the fatigue literature, are you?" If fatigue is accumulating, I don't care if the incremental >> effect of each workout is smaller, you are creating a larger effect over time. >> And say that again because I don't think people got that.

44:43 So, the problem is that you could you could have a smaller and smaller effect of each workout, but if you are still adding to what's already there, you're going to end up with a worse swelling effect because your inflammation is worse because your muscle damage is worse because you've accumulated more fatigue. >> Accumulation of fatigue isn't like, you know, kind of a computer character kind of reducing its strength. It's like you are creating more damage inside the muscle which is going to create a bigger inflammation effect which is going to create a bigger muscle swelling effect which is going to end up with a worse problem than you had at the you know earlier in the training program even though maybe each workout is producing slightly less damage yourself. You're still accumulating. That's the issue.

45:28 >> Yeah. Okay. Yeah. I look I feel like we could go down that path a lot further and maybe we shouldn't take that detour. Let's jump back to measuring the actual swelling. So Cody, you said use bio impedance. Is that the best way or is that a reasonable way to actually do this? >> Well, as I mean again, as I say, I'm not I'm not the kind of most creative person for inventing kind of protocols for you know, figuring out how to bypass limitations like the one that we're facing. I mean like where we are is that our current hypertrophy measurements, our standard ones don't really do a very good job of estimating the component of water because it's like well we're literally measuring a geometrical change.

46:23 >> Okay. So if I'm measuring a geometrical change and water is creating a geometrical change, how do you want me to control for that? You can't. You literally can't. It can't be done. So you what you've got not not in the not in the single measurement you've got to like find a way to like I was saying you either do a a muscle swelling estimate for each person at the beginning of the training program and see whether that helps you control for it. You do maybe another one pre and postpart workout but again doesn't help you with accumulated PE. If you want to do silly volumes then you're going to end up with accumulated PE.

46:54 >> >> if your goal if your goal were not to so obviously what you're talking about there is ways to make the hypertrophy results more accurate. Yeah. Without the influence of swelling. Yeah. >> If if the goal of the study was actually just to look at swelling, then what would be the best way to just assess the swelling? Is it is does that change it or would it still be the same thing? >> You just you just have a wash out period. You just ask people not train for two weeks and then you and then you do a and then you do a single workout to see what happens.

47:24 >> Okay. Yeah. Okay. So So ultimately you're still kind of coming back to the same thing. So you're saying measuring extra intracellular water via bio impedance is it's a starting point, but that's not going to be completely the full picture. Is that ultimately where you're getting at? >> Well, yeah. I mean, because it doesn't it doesn't actually give you it doesn't give you a adjustment to your to your geometrical measurement. It's it's basically just giving you a mass measurement >> which you then deduct from your lean mass calculations.

48:02 >> So it it's not giving it's not giving you anything on the geometry side. It's literally just saying this is your mass number and then you deduct the water mass and that gives you a an estimate of what your nonwater mass change was. >> Okay? So you could use it to compare different conditions and see which condition had more swelling, but you're not going to be able to use it to deduce. >> Well, you're stuck with lean mass on you, which is not amazing.

48:28 >> Yeah. Yeah. Yeah. So, yeah, in terms of actually how much muscle growth there was, it's not going to be super helpful. But if you just wanted to see what condition caused swelling, it'll it'll give you some insight. Well, you could you can use it for what or used it for, which is to try and figure out whether you're actually gaining any muscle mass with this sort of really high volume training program or not. >> Yeah. Yeah. Okay. Anything? It sounds like the swelling piece is is a little bit complicated, like there's not a a >> Well, I think it makes a really really big mess of our current hypertroy data.

49:02 and yeah, I mean, I'm not I'm not a I'm not really in a position to give a solution to it other than, you know, we probably want to take it a bit more seriously than we are doing at the moment. And as I say, what I keep seeing people doing is just brushing it off and say, "Oh, no, it's not a problem." Well, now clearly it is a problem cuz >> I mean the disappointing thing for me is that I don't know of any studies that are measuring after 72 hours.

49:33 Like as far as I can tell, they're all measuring between 48 and 72 hours. And if if that's the case and we're saying, well, there could be over 5% increases still potentially at 72 hours. For me, it just feels like, well, okay, we've we've done this job at looking at what the hypertrophy measurements are and understanding how to understand, you know, those even tests, but why does any of it really matter if we don't have any confidence based on potential swelling?

50:00 And I think that's what why we got a lot of resistance to it at the moment because nobody really wants to kind of take a step back and go, hang on a minute. we could have a serious problem with our confidence in the data that we've collected to date. Nobody wants to do that. at some point somebody will but yeah I think at the moment people are quite happy just to carry on you know hoping for the best.

50:26 >> So moving on from swelling what would be the next would we look at recovery like that's sort of linked >> I think probably we'd want to look at recruitment. I mean, I think hypertrophy for me, you know, you kind of you we've talked about the measurements that go into hypertrophy, but what are the things that actually allow you to trigger that hypertrophy? Going to probably naturally start with recruitment. So, we say recruitment, but it's almost never measured.

50:54 there are some kind of neurohysiology studies that do measure things that get pretty close to recruitment. but generally speaking, the closest we're going to see in a practical context is going to be voluntary activation, which is there's a there's a range of different calculations, but basically they all kind of work on the lines of doing a maximal contraction and then applying electrical stimulus to the muscle both at rest and during that maximal contraction, looking at the difference.

51:27 And that kind of gives you an idea if there's any unactivated muscle fibers in your muscle when you're producing a maximal contraction. so that is not really recruitment because you could argue that there's a firing rate component. I tend to think that firing rate is massively overrated. I think that it's a lot smaller than most people think it is as a component of muscle force in an isometric contraction. but you know, that's my perspective. so yeah, so if you got voluntary activation, that gives you an idea of what your voluntary activation deficit could be. again I posted recently about this trying to get people to understand that if they see a voluntary activation deficit in the literature of 90% or 95% or like the biceps you tend to see like 98%. people are like oh no look there's no voluntary activation deficit where it's 2% you can just discard that.

52:16 Well, not really because what you've measured that where you've measured that the context you measured that is a maximally stable exercise where you're strapped into a dynamometer and you've got u a very small amount of muscle mass that is working. So you've got you know kind of just a single arm and it's a biceps curl and also you're doing it in an instantaneous unfatigued state. Okay. So none of those three things are true at the end of a strength training set to muscular failure. So if you're saying to me that because a dynamometer study found that the biceps have 98% body activation, you think you're going to get 98% activation in your 10 rep max biceps curl standing in front of the mirror. No, you're not.

53:05 You're never going to be anywhere close to that because you got much less stability. So your equipment comes down. you've got more muscle mass because you're doing two arms at the same time. and you're in a fatigue state because you've just done like 10 reps. So all those three things are going to pull down your voluntary activation level, you're not going to have the same level of activation you would do in a in a dynamometry situation. So when you see like biceps up at like 98%, you think you can just say, "Oh, no, I'm maximally recruited."

53:34 No, you're not. if you see quadriceps down at 85% or even lower, it's probably even lower than that in a scenario where you again got those variables of a strength training workout. >> Okay. So the way it's being measured in those conditions, it works. It's accurate, but you're saying when we apply that to the gym, there's going to be an extra >> definition. The same thing. >> Yeah. Are there are there ways it's it can be measured in a standard strength training context? It's always No.

54:07 >> No. You're going to have to. >> So, we don't actually know what that variance could be. >> Not really. No. No. We just have isolated those specific variables in other in other experimental context. So, we know those factors matter. >> Yeah. Okay. most of the studies like this is really going to be for looking at voluntary deficits like really because you can only do that in very specific scenarios because of the equipment. >> the rest of the time you're going to be looking at proxies for activation.

54:42 so you're going to be looking at EMG could be surface. It could be fine wire. It could be high density. you could be looking at you could be looking at T2 signal intensity with MRI. You could even looking at post immediate post strength training set swelling because as I said before in a in the context of a workout the swelling isn't caused by inflammation and damage. It's caused by hydrogen ions. So if you have a high hydrogen ion content of your muscle fiber and it pulls water in by osmosis, then you're going to end up with a swollen muscle immediately post set. is going to tell you that those fibers in that muscle have been kind of working pretty hard. So, it's a way of measuring as proxy of measuring activation. So, the reality is T2 signal intensity is giving you a very similar out outcome measure. It's measuring pretty much the same thing.

55:37 and then really the only other one is mechanomography which is kind of the vibration of the fibers. So, generally speaking, most people going to look at EMG as the primary one. 90% of the time they're going to use that as their activation measure. but you can use any of those. I mean it'll all give you more or less the same answer. >> And okay, and how well are these actually telling you about muscle activation? Well, so this is one of those kind of points where it's starting to again get to a it's going to require you to have context to understand what it is that you are able to measure and what it is that it's pointless measuring. So talk about EMG for a moment. What you'll find is that if you spend any time in any comment section of any kind of popular social media kind of influencer, you'll come across somebody who is repeating the line that EMG doesn't correlate with iert. Therefore, it can therefore you can ignore it. And it's almost like one of those kind of knee-jerk kind of reactions or like every every time I kind of kind of hear somebody say it, it reminds me of Have you seen the the the John Candy Steve Martin film Planes, Trains, and Automobiles?

56:58 >> I have not. >> You haven't? Okay. It's a bit of a Gen X film, I guess. and I think it's that film. I think I remember that film correctly, but there's definitely a scene with there's definitely a a film with John Candy talking to Steve Martin and they're playing opposite each other. I'm pretty sure it's Planes, Strange, and Automobiles. And he basically Steve Martin basically describes John Candy's character as being one of those kind of small toys with a a kind of a string in its chest that you can kind of pull and it'll start talking. Again, Gen X, I'm describing something that you guys probably never ever come across in your lives before, but there used to be these toys that you could pull a string and it would make a noise. And he basically was accusing John Candi's character of being one of these toys that you kind of it would pull its own string and it would just make a noise all the time. And it's like that's what these kind of guys are when they go, "Oh, EMG doesn't correlate with hypert." It's like they pull a string, EMG doesn't correlate with hypert. That's all they can say. It's this they don't know why they're saying it and they don't know where they got the information from. but they know that it's true. They know that it's true. They've got this emotional belief that it's true. And it's like, where do we get that idea from? We got the idea basically from from from the observation that you can get really messy EMG results from comparing different states of fatigue. So, if you have an EMG output from two different fatiguing contractions and one's got much more fatigue than the other, you're not really going to be able to compare them very very effectively because fatigue messes up your outputs. If you're not doing that, if you're maintaining kind of a relatively low level of fatigue or keeping fatigue pretty much the same, then you can start to compare those scenarios. But what are you actually comparing them for? You know, what are you actually trying to do? I think that essentially in an unpatigued state, what EMG gives you is a pretty good proxy for the number of muscle fibers that are activated, more or less. So we can do that through a number of different steps but fundamentally where we're kind of working through is that if you have a isometric kind of contraction and you take EMG measurements and force measurements every kind of 10 20% of maximum force and you pretty much get a straight line of EMG and and force obviously and that's because as you increase the number of fibers are activated you're increasing the force production and that simultaneously increases the number of fibers that are registering as having a EMG signal and that's what you pick up on your surface electrodes assuming that they're correctly placed.

59:36 So broadly speaking you're getting an idea of how many fibers are activated. Now that as I say messes up if you start kind of comparing different fatigue contractions. So if I have one scenario which is a lighter load but I keep it active for a longer period of time and I start getting an increase in motine accumment my EMG signal is not going to reflect that in the way that it would do if I was doing it in a precise unfatigued state with you know kind of different gradations of force and allowing rest in between each of those tests. Does that make sense?

60:09 >> Yeah. Why is that? >> Why is that? Okay. no, no, we're not going down that rabbit hole. it is I mean it's basically it's it's a is a is a seriously confounding issue that we face when we're trying to measure EMG is that fatigue will kind of create problems for us >> to the extent that it's like entirely unusable or to the extent that it just isn't it? >> It's going to mislead you. It's going to mislead you, man. It's going to mislead you. so yeah. and and that's that's that's that's me kind of identifying the thing that's most important to me because I'm already discarding stuff that that's going to trip other people up. So, if you I mean, I mentioned this recently. If you're measuring EMG in a scenario where the muscle fibers are not shortening slowly, then you're not measuring anything that's going to be remotely related to muscle growth because you obviously got got single fiber tension high enough. So you can measure you can measure you can even if you're like you know I'm assuming that you kind of controlling for this stuff but if you're not controlling for that kind of stuff you're going to have an extra set of problems as well. And that thing that's probably where people get to when they try and go, you know, oh, can I correlate EMG with with with hypertrophy after a long-term training program?

61:35 Well, you're measuring integrated EMG over the course of a whole set, then no, it depends on what you're doing. >> So, in what way do you think it's it's best used? So, you said I think you said it. >> I think the only valid use, not valid, that's probably a strong word. The only sensible use for EMG is to compare activations of different muscles in the same strength training exercise. So if you've got a normalized kind of EMG, so you've taken EMG in a maximal contraction of the muscle is actually working really strongly and then you kind of do that for each muscle that you're going to test and then you slap EMG electrodes on all of the muscles and you get someone to do the exercise, whichever one gets closest to its maximum limit is probably the one that is the muscle that is going to be most developed after that exercise in a strength training program.

62:26 >> Okay. So its best use would be relative. So it's not going to necessarily tell us. Yeah. >> Yeah. >> Yeah. So so we can't use it as a proxy for like what percentage of that muscle is being activated. >> Well, as I say in the context of a isometric or a relatively unfatigued contraction, then yeah, it does give you a reasonable proxy for what percent muscle is activated. >> >> But you know, you've got to be clear about what you're comparing things with because if you have a dynamometer scenario or an isometric highly stable scenario and you get your 100% in that context and then you test a dynamic contraction, you've got two problems.

63:07 Firstly, you've got a problem because your stability is now lower. So, your recruitment is going to be lower no matter what you do. Secondly, you've got a dynamic contraction. So, your firing rates are going to be higher. So you your EMG is going to be higher than you expect it to be. So if you go and if you move too quickly, you're going to find that you blow past your normalization point. So classic scenario, this causes all kinds of aggravation in in SNC, not so much in hypertrophy sort of studies, but in SNC, what they'll do is they'll do an isometric contraction test. and they'll normalize the EMG to that.

63:41 and then they'll do a dynamic contraction like a power clean or something fast and they'll go straight past their 100% normalization value. They'll be if they go like do a jump or something, they could be at like 150%. >> You know, the sprinting studies with EMG values way off the top end of normalization value. Why is that? Because firing rates are way faster in fast contractions than they are in in slow contractions. That's got nothing to do with the recruitment level.

64:08 That's just how how how how the firing rates work. >> So if you're using EMG and you're trying to work out, okay, which which part of this muscle, let's say we're looking at elbow flexes, trying to work out which elbow flex is more activated in a movement. >> Well, straight away there, your problem is you're going to have crosscontamination because EMG is going to pick up anything in its vicinity and you can get a lot of cross talk. This is one of the problems we've got when you're trying to separate out recam from the other quadriceps. A lot of the time the rec will pick up as active even when it's probably not.

64:44 >> Mhm. Yeah. Okay. So, so you're saying then it it may not actually do a good job at that. So, so where >> I'm lost in >> No, no, no, no. I'm I'm kind of like imagine in a squat and you slap electrodes on the glutes and the adductor magnus and >> you know kind of the quads and the calves and whatever. You can you can kind of see >> okay so you can compare the the quads to the glutes >> and that would be using that would be using sort of basic surface EMG. You want to kind of start to get granular about muscles within the quadriceps. You can either need fine wire or you're going to need high density.

65:21 >> Okay. So that's important because often when people will just see the word EMG, they may not realize that there's different types of EMG. Yes. >> So you've just said, okay, we got surface, fine, wire, >> 90% of it is going to be surface. >> Okay. So can you just touch on the differences there briefly or how you would what they're going to tell you? >> So surface, you're going to sort of get whatever's underneath the surface of your electrode.

65:46 >> Exactly as it sounds. So it could, you know, it the problem is it's a little bit kind of regional in that regard, but you can also get contamination from stuff around it, especially if you got very little going on underneath your exact electrode. >> Okay. So, so surface will not be great for within muscle differences. >> Within muscle group differences, yeah, >> not really going to be very helpful. high density is going to start to address that because you can see trends, but again, it's going to be a bit blurry. but it is going to give you I think there's really I think there was a really nice glute study done recently with high density MG that showed basically kind of everything that we've been saying before about regions and kind of differences in activation between regions like you know abduction and kind of external rotation and extension.

66:37 >> so yeah sort of high density is where you're going to see most of the new research coming out. Fine wire is old studies. You're not going to find a lot of that except in the physio literature because you kind of need a doctor to be present to stick the needles in people. So, you're generally not going to be able to see that done very commonly cuz it's just, you know, kind of requires sort of other people to help you.

66:59 >> Okay. Okay. So, how much of it is going to be influenced by where the actual electrodes are placed? Like, is that something that people need to consider? Well, that's that's basically the core of a a disagreement that was kind of breaking the internet a while back, a year or two ago on the trapezius because unfortunately the way that surface EMG guidelines work is that everyone uses the same guidelines and if there's a error in the guideline then everybody uses it. So unfortunately when the guidelines were constructed for the trapezius the guidelines stated that the upper trapezius was located at if I get this correct I think the error was it was located at C7 and in fact it's located at C6 and above. So if you put it on C7 you can get in the middle trapezius instead of the upper trapezius. So a lot of the EMG studies on the upper trapezius actually measure at C7 and therefore our studies measuring the middle trapezius.

68:03 >> So again that is something that nobody is going to know to check unless they get top. I mean that's just fortunately that's the problem we've got. That's how things work. >> Okay. So, as far as actually looking at within muscle group differences, just saying that if it's just on using surface EMG, probably not going to be super helpful, but if you're using fine wire or I mean, I wouldn't go so far as to discount studies that are using surface EMG within a muscle group. I think the the delt studies are not too bad when they measure, you know, kind of anterior, middle, and and posterior. I don't think they're horrible. when they use surface EMG, I think they there is contamination from cross talk there, but I don't think it's it doesn't seem to necessarily produce silly results all the time.

69:00 >> Okay. >> but I think it's definitely a concern that that is that is a risk. >> Okay. Okay. So, it's worth being aware of and you need to Yeah, I guess consider it. Okay. anything else to do with activation or is that >> I mean I would just mention that most of the activation methods when they're done in an unfatigued state tend to cross cororrelate with each other pretty well and we tend to crossorrelate with muscle force pretty well. So you know as you linearly increase muscle force in an unfatigued state in isometric contraction. So you do repeated tests at different force levels. you can get a pretty linear relationship with EMG measurements with mechanomography with T2 signal intensity with you know I mean ultimately they they all kind of give you more or less the same kind of gradation. The issues occur when you start to have differences in fatigue.

70:01 you know, you're going to start to see that breaking down. It's not going to work anymore. >> So, is there a way to actually then get good data on muscle activation in more fatigued or fatiguing states? >> Well, it's not going to tell you anything useful. I don't think I don't I genuinely don't think it's the EMG is going to tell you anything useful. I mean, I think >> yeah, I I really don't. and the I mean you can start looking at you can start looking at decomposition algorithms and try and figure out what's going on using those. but honestly there's a lot of trust involved in what you think those are telling you.

70:45 >> Mhm. So, yeah, I think it's it's kind of you you kind of get I think you get to a a cleaner place by just kind of making some assumptions based on basic physiology and then kind of not not getting buried too much in the detail of those of those kind of stuff. >> Okay. So that statement that you led with when people end up in chat forums and say EMG doesn't produce hypertrophy ultimately you're saying yeah in in certain circumstances where it's perhaps applied or or misapplied then that's true but we can probably use it as a decent proxy for what's going to grow if it's used in an unfatigued state.

71:30 >> Yeah. I think if you've got a a study that's just slapped a bunch of electrodes on people and asked them to do a couple of reps of an exercise, then whichever muscle is highly activated compared to the others is pretty much going to be the muscle that's going to grow. And that's literally the only way I use EMG. >> So when you see on on social media there's accounts who you know they'll have an AMG device and they'll do that and they'll have a participant or two and they'll do different exercises or whatever and they'll you know compare them. Would you is that being done in a way that you think would be helpful?

72:03 >> So the issue is that generally speaking you're going to want particip you're going to want a decent number of participants. You can't just slap electrodes on your three best buddies and kind of get, you know, data that's going to be meaningful. You're going to want like 20 odd people really. but yeah, I mean like if they are putting the electrodes on the same exact place every single time and it's the right place and you're normalizing for a maximum contraction level because what you can't do is just take the the number that pops up on the screen. You can't do that. You got to know how what is that relative to your maximum contraction level. And that means you got to have the exact same placement of the electrode. Ideally, you're going to do this same this in the same testing session. So, you slap the electrode on right place and then you get a position that produces max recruitment in that muscle and you can there are tables of those kind of particular joint angle combinations all over the place. I mean, you can get them quite easily and produce a max effort in that isometrically and record your EMG value for that. And you got to decide what you going to measure. Are you going to measure the peak or are you going to measure like an average or are you going to measure like an integrated value and what your equipment is capable of doing? And what can >> peak not be influenced by fatigue?

73:30 >> Well, we're talking about doing these in unfatig scenarios. >> Yeah. Okay. So, how what what constitutes an unfatigue scenario? Because if you do a set to failure, presumably there's going to be fatigue. >> Yeah. You're talking about doing a couple of reps with high effort basically. But that means you're going to need to be sort of at the heavier end of the spectrum because you're not going to want to be doing, you know, kind of really fast repetitions cuz as I say, that'll blow past your normalization. It's a bit of a mockery of what you're trying to do.

73:56 >> But you're going to need a normalization and you're going to need to decide what is it that you're normalizing. Are you normalizing? I mean, I have opinions about this stuff and I don't want to kind of derail this and turn this into a methodological argument about how to use EMG, but you are going to need to make some decisions about what you think a normalization should look like. I mean, for me, I think you only really want to be normalizing. If you're going to use peak, then that's probably the easiest one to normalize to because you just take the peak of both numbers. If you want to do average well okay mean they normally call that well okay what do you which slice of your normalization section are you taking and which slice of your exercise test are you taking okay and if you want to use integrated the same problem app flies but it's now a bigger problem I quite like peak for the for the kind of fact that it probably does away with a lot of those issues I think you can get errors from using mean or integrated that shouldn't be there just because you taking the wrong slice.

75:03 >> Okay. When I see I mean this is just social media, but when I see these accounts pop up, typically I think what I'm seeing is they're just taking a set to failure and then they're comparing that to another exercise, taking a set to failure. >> No, no, no, no, no. You can't do that because the velocities will be different. you know, so you know, if you if you're using a different range of motion, then the veloc velocity is likely to be different across those two exercises. and the velocity is going to affect firing rate which can affect DMG.

75:31 I don't like bearing exercises so much as bearing muscles within exercises. I mean, I think it's not horrible if you've got like sort of a decent proxy for an under fatigue state like you're doing a couple of heavy reps. I don't think it's horrible, especially if you're doing similar ranges of motion. Like if you're comparing two different squat variations, I'm not like massively opposed to that idea. I think that's probably reasonable. You're comparing two different pressing exercises with similar ranges of motion. But again, I'm not like I don't that's horrible.

75:59 >> So, what about if you're like changing hand position or something like that with the same type of exercise? Would that be >> Again, I'm not not going to problem with that. But I think like for example, the classic I mean, fortunately, it doesn't make that much difference. But, I think the classic comparison of the hip thrust versus the squat massively disadvantages the hip thrust because because you're using a faster velocity in the squat than you are in the air thrust. Right. Okay. Okay. So, the main issue you have is is due to the velocity change.

76:31 >> Well, it is the it's the one that's going to create the biggest effect. >> Okay. Okay. See, this is interesting because again, like I see this stuff thrown up on Instagram and people be like, "Look, here's a, you know, this guy's got an EMG or someone whatever and then look at this." And it's like, "Yeah, but does that mean anything? What are you looking at?" And >> the f the first issue is that it needs to be >> like a if if if there's a normalization happening and there's and and there's a very clear calculation for how you're doing either the peak the mean or the integrated and that makes sense then you know that's a really really good start and if doing those kind of things and they are going to the trouble of of of controlling for normaliz you know normalizing the whatever they're to a a max effort isometric contraction EMG value, then that's a great start. And as I say, if they're doing a very clear calculation of how they're getting their EMG amplitudes or whatever, again, that's a really a good follow on. But where they're going to fall down, even if they are doing those things, is that they probably are going to have a couple of people at best having a measurements taken. What you really need is probably, you know, K 10 to 20 >> just because there's that much individual variability or >> because otherwise you're going to end up with a really high noise to signal ratio. What you want is a bigger signal to noise ratio.

77:56 >> Okay. So, I feel like EMG could be a whole episode in and of itself and I'm aware we've talked about >> but as I say, I've got really strong opinions about how this stuff should be done and they don't always line up perfectly with EMG researches. So, I've upset probably just about everybody in the on the inert at this point apart from the EMG researchers. So, trying to keep it >> Okay. So, so let's let's keep them intact and let's not upset them. So, what that's it for activation I forget.

78:27 Did we say there was >> yeah, I mean basically you've got kind of you've got two questions you can reasonably answer. You've got what's the voluntary activation deficit of a muscle and you >> which you're saying we can't particularly answer in the context of a gym setting. >> You can you can get a you can get a benchmark number for a stable setting and then you can kind of backtrack it a bit to get to what you're likely to find in a gym setting. Qualitatively you can work backwards from then and see it's going to be worse.

79:01 >> if you were to hazardly guess how much worse, like if we're saying biceps might be 97, 98, whatever percent in the gym setting, how much how many percentage are we expecting to drop? >> Well, it's going to vary whether you're using machines or whether you're using barbells or whether you're using dumbbells. >> Sure. Let's say you're using a preacher curl machine, something that's fairly >> as you're not going to be miles away. I mean, if you're doing single arm preacher curls on a preacher curl machine, you're not actually going to be miles away from what you get on a dynamometer. you're only going to have that lack of stability of the rest of the body. If you're a big guy and you're wedged into the machine, it's not going to be that different, you know, like me, I'm all over the place.

79:41 >> A leg extension presumably then similar. >> Same issue. >> Yeah, >> same issue. As soon as you start moving towards like free weights and and like cables and stuff like that, then you're going to be a long way away from, >> you know, >> you know, >> if you want to really start to nail down some of those numbers, what you would need to do is look at the actual performance differences. if you can make sure that your angle if make sure that your axis of rotation is identical in each of the cases. So, if you've got like a dynamometer, you can get a torque output for that or a force output for that.

80:22 Look at the same thing on a a preacher machine. Look at the same thing on a a preacher curl bench with a dumbbell and see what your force output differences or torque output differences are in each of those situations. the difference is going to be purely motivated because you're going to lose stability as you go down that that that that >> so so are you saying there force output can be used as a measurement >> in that scenario you could use force output to estimate what your differences are in in recruitment because you've controlled for everything else I mean you could argue maybe there's kind of coordination issues but if you're familiar with the exercises that shouldn't be too much of an issue >> is that used often This this is not this is not an acceptable you this is not an acceptable measurement method. So this was not on my list. this is me doing what I do which is trying to figure out based on data that exists.

81:21 >> You're saying force output is not not an acceptable measure. Yeah. I was thinking is I don't know I see that. >> Okay. But force output is measured. >> And does it tell you anything? I mean this is the fascinating thing because we we have like a billion measurements of strength and force and stuff across the literature and technically that is a pretty good proxy for whole muscle tension and yet it tells us nothing about hypertrophy >> which is one of the kind of really big hints that whole muscle tension is not a stimulus of hypertrophy but >> I guess I'm biased on that one. So in what what context what when would you come across force output in a in a study? What are they usually applying that to?

82:08 >> Every time you're measuring strength, you're measuring force, aren't you? So I mean >> you're going to when when are you going to see that? I guess you'll see that in like >> yes strength change measurement somewhere in there. >> you're not going to see this in the hypertrophy section. No, you'll see it in the in the strength section >> and as a proxy for strength change. This is helpful. >> Well, no, it isn't a proxy. It is actually the strength.

82:38 >> It is. It's what's measured. Yeah. So, are there things to consider? >> But I'm saying like I'm saying that you know we moved from hypertrophy to recruitment because we're talking about you know the most one of the kind of inputs for hypertrophy is the recruitment level. but then I'm saying whole muscle force is literally muscle tension. So why doesn't that get used in the hypertrophy kind of literature is something useful? Why why if we use recruitment all the time to tell us about strength train? Like why would you go to the trouble of measuring recruitment or proxies of recruitment of activation if what you really want to know is tension?

83:19 It's a clear kind of disconnect in the literature when you've got hypertrose researchers saying that muscle tension is what they think is driving hypertroin. Like well why aren't you measuring it then? >> Why are you measuring recruitment or activation? Why aren't you measuring muscle tension which is literally just force? >> It's because it doesn't do anything. It doesn't tell you anything. I mean the classic classic metabolic stress study, you know, where they literally controlled for force time integral across the two different groups and they found that one group achieved hypertroy because they were doing all of that force time integral in short period of time creating >> basically increasing motion as a result of the fatigue that was accumulating and the other group was doing it in spaced kind of chunks with rest. They didn't achieve any hypertrophy. The fourth time interval is identical in both training groups.

84:10 >> So, you know, you're controlling for muscle tension, but you're not. So, you know, like I don't understand why hypertrophies keep hypertrophy researchers keep coming back to this idea of muscle tension being a driver of hypertrophy because it can't be because we literally got studies that have discounted that, you know. >> So, it seems to be a disconnect. So, if they did believe that, then they would be using force. But >> exactly. So the behavior force is telling you everything you need to know, >> you know, and I said this before about human beings. Don't don't listen to what they say. Look at what they do. and hypers don't measure muscle force.

84:42 >> I think I think Mother Teresa came came to that same conclusion. I'm pretty sure she said don't don't listen to what people say. Just judge them by their actions. Yeah. >> but in hypertrophy researchers don't measure muscle tension. So in like muscle force. So why would why would you take them seriously when they think it it drives hypertrophy? ally doesn't it's single fiber tension obviously that drives and you can't measure that in humans >> yeah sure whole muscle force are there like if are the things to consider when we're looking at a strength study and I know you we're talking about hypertrophy mainly but is that that is a a >> is it easy to stuff that up or is that like no great this is a a a safe measurement we can have confidence when we're seeing whole muscle force >> it's phenomenally easy to stuff it up I mean basically the the the primary way that they stuff it up is having a disconnect between the measurement method and the training study.

85:36 >> Yeah. Sure. Okay. >> and there's some classic examples of that. And some of them actually like I mean one of my very favorite examples of this exact phenomenon was instrumental in kicking off a little niche area of study. So the stuffing up actually was really useful. Yeah, basically they did a leg press strength training study and they used they wanted a non-specific strength measurement rather than just measuring leg press strength gains. They wanted a non-specific leg strength measurement. So they use knee extension isometric force. no problem with that. If you want to kind of non-specifically assess quadricep strength as well to leg press training, that is exactly what you would do.

86:21 But what they did was they used untrained subjects which again not a problem but they used untrained subjects and they measured the isometric knee extension force relatively close to full knee extension. Now the problem with that is that if you do leg press training with untrained subjects you are going to get pretty meaningful increases in cycogenesis in your quadriceps and that is going to shift your length tension relationship. as a result of shifting your length into relationship, you are now going to find that you get weaker in the extended knee that's stronger in the stretch position. And so what happened in this famous study, famous at least in this kind of niche area that I'm referring to >> is that they they didn't display any gain in knee extension strength at this position even though they got a massive increase in leg press strength. And to start with with they're like puzzling about it. They're going it's really just telling us that we've got massive coordination gains in the leg press or or has something weird happened and it actually they then did a modeling study and they they went into soenesis literature and they ended up being able to show that it was purely so genesis that had shifted learn tension relationship and basically you've simultaneously decreased your ability to produce force at this particular joint angle while simultaneously increasing strength across the board. really cool example of how an error actually opened up because >> got inquisitive about it. They didn't try to sweep it under the carpet like the evidence based guys do. They just go, "Oh, no, no, that's an anomaly or that." You just ignore that. Like instead, they got inquisitive and did a science-based kind of approach and what's it telling me?

87:55 >> And they actually tell tells us something pretty cool. >> Yeah. >> Yeah. You can't stop showing measurements if you don't know what you're doing about like what's happening around the like context is everything. >> Yeah. I guess it comes back to what we were saying about the exercises that are being used and the muscles are being measured. I guess again it has to has to make sense for what the the actual study was methodology was doing. okay.

88:23 Okay. I don't want to talk more about strength because this is hypertrophy. So what's left? recovery. >> Well, really segueing straight from strength into recovery. And I mean like basically we use strength as our primary recovery measurement proxy. So when and this is I've said this a million times when people talk about recovery in their minds. I think they've got this little computer sprite character who's got a state of recovery bar that's kind of ticking up to 100%. they don't or they think about how they feel or they think about whether they're ready or whatever. And that's just a million miles away from what research is actually doing. What we're measuring is has strength recovered back from your previous workout. That's literally all we're doing.

89:13 >> Subjective like that will be a measure that's used in some of these studies though. >> You can record it. You can record but it's not the standard. The standard is exercise performance because fatigue is not defined subjectively. Fatigue is what we refer to as being absent when you've recovered. So the process of recovery is the dissipation of fatigue. >> You want to see it? And how do we measure fatigue? We measure it as a temporary reduction in exercise performance as a result of a previous bout of exercise. So it's like >> how's strength got back to normal again?

89:44 Well, okay. >> Like I don't feel recovered. Well, I don't care because >> Yeah. So we can't use subjective feeling as a proxy for recovery. >> Well, it's not. I mean, you can kind of talk about it. You can measure it and you can talk about it and you can say it's interesting but it never interestingly never goes anywhere. Nobody ever goes anywhere with it. But if it correlated perfectly like you could use it as a proxy but it doesn't even correlate does it?

90:07 >> Okay. So subjective well-being pain doms that would be in that same category. That also doesn't correlate well enough. Yum. >> Yeah. It's I mean it's again it's one of those kind of measurements that people take and they kind of talk about but it never goes anywhere because it's not actually the thing we're measuring. >> Yeah. So it's just Are there any other proxies or measurements that that do correlate fairly well with strength recovery? >> I guess there's not, is there?

90:39 >> No. And you have to bear in mind that what you're what you're actually trying to get at is a basket of fatigue mechanisms that are under the surface. So you know fatigue is not a thing. It's a it's a measurement. >> Recovery is not a thing. It's a measurement. So what you're measuring is a strength change. >> now what's causing that is a set of fat mechanisms. Those mechanisms can be central nervous system, they can be peripheral and they will do different things. And that's kind of where it starts to get interesting because if you have a motiv, that's going to affect the whole body.

91:23 Whereas if you've got just a peripheral muscle damage effect, it's only going to affect the muscle that you've trained. So it starts to change the way we interpret training guidance based on you know kind of training splits and postwork fatigue and when you're going to train again and all that kind of thing. So it starts to become very practically relevant quite quickly. But ultimately the measurement method we use is also going to be different have differences in sensitivity because if you look at the way a muscle is constructed you've got element size principle top end you've got the kind of high threshold motor units and a lot of those are going to be connected to or you know contain fast fibers and down the bottom end you're going to have more oxidative fibers.

92:07 So ultimately if you perform a isometric contraction broadly speaking there's going to be very little difference in the amount of force that a fast twitch fiber and a slow twitch fiber are going to be capable of producing for the same cross-sectional area going be 10 15% difference max in contrast if I do a high velocity contraction the slow twitch fibers are basically not going to contribute at all and your fast twitch fibers are going to contribute everything so if I test my recovery based on isometric strength which is a common process I will basically be saying you know let's imagine that 5% of my fibers got damaged in a particular exercise and just for cleanness of calculations let's assume they're going to produce no force not going to work like that but let's just so I'm going to have a 95% strength output instead of 100% strength output from my isometric >> contraction >> if I now do a vertical jump and I my fast my slow fibers can no longer contribute. Then basically I've just chopped half the muscle off and said I can't use that bottom half of the muscle of such.

93:14 >> So now I'm taking that 5% and I'm dividing it by 50 instead of by 100. >> So I'm now going to get a 10% difference in vertical jump performance compared to my 5% difference in maximum strength on the isometric contraction. M >> so generally speaking fast movements are always going to give you a much more sensitive output for your recovery compared to your maximum strength u measurements and they're both totally valid just one's more sensitive than the other.

93:43 M >> so this is why in SNC context not the only reason why because vertical jumps are a lot easier to test than you know kind of maximum isometric strength for you know kind of anything else at that end of the spectrum but you know it's one of the reasons why we use vertical jumps to test kind of recovery because it's just so much more sensitive. Very quick disclaimer on that. If you're a bodybuilder who's thinking about doing that, take your time and give yourself kind of a month or two to get to the point where your vertical jump stops changing because if you want to use vertical jumps in your as a as a as a proxy and you can just download an app off the app store and kind of start, you know, testing your vertical jump height very easily just by putting on a, you know, giving it to someone to kind of point at you where you do a vertical jump. But you will find that your vertical jump actually moves quite a lot over the first couple of months of just jumping.

94:35 >> Is that going to be the same for for both? Like even if you're doing an isometric then because often you will see in some of these studies that that they'll actually be above baseline a few days later. >> Yeah. Yeah, I mean you could have an adaption, but generally speaking, yeah, if you've got untrained people, you can find that the coordination effects do kind of mess things up, but much less so on the isometric side and the dynamometry side is trapped in. There's very little room for coordination get and they tend to do a bit of familiarization processes as well to try and avoid that happening. But vertical jump, you can't do that. It's just going to it's going to take you weeks to get to a point where it stabilizes if you've not been vertical jumping for a long time already. So yeah, if you're a bodybuilder and you not done any sport for a while, then you know, give yourself a couple of months before you actually start using vertical jump height as a you know, just practice it a bit. and take your time. cuz it's just going to keep going up. I mean, literally, it's keep going up.

95:31 Like I've I've I've seen I've seen people do a vertical jump testing session and we can't get to the point where they're getting worse. It's just >> because the learning effect just keeps going and you just give up. It's like It's just pointless at this point. You know, you just got >> So in studies when they're using vertical jump, there's enough of a familiarization phase that it is actually going to be a helpful measure. >> That's why I tend to try and prefer vertical jumps in athletes because you don't get that problem.

96:00 >> Yeah. Okay. Interesting. Okay. Are they the main main ways that recovery is actually being measured? So you got the isometric force, you got the vertical jump. today any other >> well you know you can start to try and get into the data of which fatigue mechanism you're measuring so you can measure voluntary activation so you can measure changes in motor recruitment you can see super spinal C being present in the post-workout period you can measure low frequency fatigue which is basically a proxy for all of your calcium related fat mechanisms I mean those >> I don't think I've seen that in a study No, you have. No, you don't.

96:45 >> Is that what it's called? Low frequency. >> Yes, it will get referenced as excitation contraction coupling failure. >> Okay, >> that's what it will get referenced as. It's not. It's low frequency fatigue. So, basically it is two electrical twitches, one at either 50 or 100 hertz, one somewhere between 1 and 10 hertz, and they take the ratio of the low over the high frequency kind of number. And basically what you'll see is that low frequency force tends to decrease more than high frequency force and that is an indication that you've got a a calcium related D mechanism present. The assumption is that during the workout you might get some cyclal excitability reductions because of cell membrane damage. You might get some losses of myofibrill sensitivity and you might get some excitation contraction coupling failure.

97:37 So you might get all three of those things. The assumption is that the cell membranes repair themselves really quickly which they do with a caveat. And the assumption is also that my sensitivity goes away which it does again with a caveat. And because of that, they're going to go, "Oh, okay. If I've got if I'm measuring low frequency fatigue 24 hours post-workout, then because my cell membrane should have repaired themselves, and because my or my, you know, depolarization should have fixed itself or my fiber sensitivity should have gone back to normal. I can assume that everything is just excitation failure, which is you can see what they're trying to do. The issue is that oxidative stress has a nasty habit of bringing back both of your other two calcium T mechanisms.

98:19 brings back your cell memory knows and it brings back your my sensitivity loss. So essentially you end up with all three again and you've just got a proxy for calcium D which is not a problem because they all do the same thing at the practical level which is reduce single fiber tension. So if you're interested in it perspective you can actually just say well it's the same as as if I've got just excitation congestion coming. So you're not missing anything unless you're an obsessive about fatigue literature like I am.

98:47 You're not missing anything. Yeah, practically it's still giving you the same information. Okay. Okay. I hope people aren't feeling too overwhelmed. It's a little bit messy, isn't it? >> It's horribly messy. It's absolutely horribly messy. It's really really horrendously messy. And that's that's >> that's the >> and that's the issue because everything everything has variables that matter >> and everything affects everything else.

99:21 >> Yeah. Well, that's the thing cuz you know I guess what we've tried to do here is just I I suppose explain what these different measurements actually are. And in some examples you've said how things will affect other things, but there's so many other things that will affect those things that we didn't talk about. And this is just step one. Just actually having some idea of what it is we're even looking at, let alone how all these things are are playing together and influencing each other.

99:44 >> Yeah. I mean, again, I don't want to come across as too Blackpool, but you know, honestly, I think unless you've got, you know, kind of 10 years to put into this stuff, I don't really see you going to get a lot out. >> Yeah. And, you know, we haven't even touched on anything statistics or sample size or anything like that. So, >> so yeah, I mean generally generally speaking, very brief comment on statistics. maybe two brief comments on statistics.

100:18 I agree with the position that some of the evidence-based guys take, which is that most statistical power in sports science, exercise science is on the weaker end of the spectrum. I agree with that. But equally, I think it's relatively uncommon for a study to have a to have a serious statistical problem that discounts those results in the context of other exercise science studies. So those two statements are true, but they look conflicting. So on the one hand, I'm saying that yes, generally speaking, sports science doesn't have good statistical power as a rule. Equally, I don't think there's a massive amount of variation within that category of statistical kind of sort of ways that the studies have been designed. So, I don't think like I think what tends to happen is that people will try and discount a study because they don't like the way the stats were done.

101:20 I think that's actually rarer than it gets presented as being as it gets presented as happening. And most of the time, the stats are much of a muchness. They're kind of they're not no stats in sport science is good, but generally speaking, you know, kind of they're all kind of in the same sort of bracket. I don't think you're going to find studies that you can just throw out the window because you don't like the stats. People do that, but I think they're being disingenuous.

101:47 >> Is that not the idea then behind a meta analysis where it's like, okay, maybe we don't have enough statistical power on one study by itself, but if we throw them together, then we do. >> Sure. And this is this is kind of and in principle, you know, that sounds like a great idea. The issue is the issue is that if you don't examine your priors very carefully before combining data sets to decide whether you think it is even appropriate to combine those data sets, then you could end up basically just throwing a whole bunch of nonsense together in a bucket, stirring it up, and then trying to make sense of it. And that's what happens in most.

102:26 >> So, so if you had 10 studies that were like the exact same thing was being studied the same way, they just were 10 individual studies, small studies, then okay, you could throw them together and that would be a helpful way of doing it. >> Absolutely. But the problem that you have, the problem that you have is that a study will have you know one particular exercise selection and that exercise selection may be different from your other exercise selections than your other studies and you may not be valid to compare those for reasons that are you know perfectly you know valid or your populations may be different like might be like a sedentary old person or >> well like pick an example that's close to my kind of area of interest like you know is it valid to have to have a meta analysis where you've got trained and untrained subjects if your interest is in measuring detecting the presence of of stretch media >> typogertrophy. No, it's not.

103:21 >> And your physiological prize should tell you it's not and you shouldn't do it. >> But >> ultimately, >> people want to do it because they want to throw more studies in and get more statistical power and get more impressive meta analysis. And it's like, yeah, but what you're doing there is you're you're incent. This is the problem I have with the evidence-based framework. And people will argue with me and say that's not evidence-based. I'm like, don't look at what people say, look at what they do. I'm not interested in what you think exercise what you think evidence-based practice is supposed to be about. I'm interested in what people actually do.

103:55 And what they do is they take as many studies as they can possibly fit into a meta analysis and they shoot all them in. That's what they do. Now, you can say they're not supposed to do that and I would agree with you, but they are doing it, you know, and the problem is they're not examining what their prior should be before they actually do. And therefore, they're combining studies that should be combined. And you get things like populations being combined which shouldn't be combined. You get studies which you know kind of have different exercises and those exercises fundamentally are different from each other. you know some are causing psychosis some are not you know some are training one muscle another training a different muscle. I mean this is the issue you actually can't combine those in the way that you think you can. M >> so but yeah I mean like the question was originally about stats and the point I'm making is that yes stats is not amazing in exercise science but generally speaking it's not going to be a reason to throw a study out the window.

104:51 >> Mhm. >> If you do start throwing studies out the window you're generally going to have to throw all of them out the window. >> And this this is probably my last question for you unless there's something else you wanted to cover. But one thing that I think that people may have observed, if anyone spent any time looking at studies, what they'll see is there can be what looks like quite large numerical differences in outcomes, but it won't be statistically significant.

105:19 >> Sure. >> And I find that especially interesting when it comes to something like recovery because it feels to me like a small percentage difference in recovery seems to be significant. like we're not using significant in the same way, but that seems to be important. Can you can you just quickly talk to what are we meaning when we're saying statistically significant and and compared to if you're just looking at the numerical numbers? >> You're going to get me into a lot of trouble now. Okay.

105:47 So, I'm not going to do a clear explanation of how statistical significance works in studies because I will not do a very good job of it after nearly two hours of talking about other stuff. and it it isn't really something that I teach. So, I won't do a very good job of it. And people deserve a better treatment than that because it isn't as straightforward as it looks on the surface.

106:17 Basically what I will say is that what we are doing with okay let me back up one step. The word significant is used incorrectly in the fitness industry compared to what researchers mean when they use it. Mhm. >> When I see people asking me questions and they go, "Is this significant? Is that significant? The other significant meaningless questions because they're not referring to a data set." >> Mhm.

106:47 >> You can't say, so this is a good test. If you think it's appropriate to ask someone, is the difference between you know, the hypertrophy of the glutes caused by hip thrust training and squat training, is that significant? If you think that's an appropriate question to ask someone, you don't know what the word significant means. people use it to mean meaningful. >> So if you want to ask someone if something is like meaningful, then ask use the word meaningful. Don't use the word significant because it's it's like people who refer to medial deltoid.

107:19 >> Yeah. As soon as someone uses the word medial deltoid, I'm like, okay, you're using a word that you don't understand what it means. You're >> using it because you think it's more scientific sounding than the correct word, which is middle deltoid or lateral. M so like when someone uses the word significant I'm like you're trying to use language that makes you sound like you know what you're talking about and actually it's communicating to me that you don't know what you're talking about because technically significant refers to a data set that you are doing statistical calculations upon at least in exercise science. So, if you're using it just in like in a generic question about different things, then it tells me that you've just picked up a word and you think it means meaningful and it doesn't use the word meaningful or substantial or big. Like, I don't care. Just don't use the word significant because it confuses people who actually know what stat how stats works >> and it makes them think that you are talking about something that you're not or makes them think you don't know what you're talking about. so but significance refers to like a you've done a calculation a statistical calculation on a data set and you've established that the outcome that you've measured is unlikely to be due to chance. It's likely to be due to a real effect.

108:35 That's all you're saying. It doesn't do anything about the size of the effect. It just tells you that it's likely to be a real thing rather than something that was just, you know, kind of random. That's it. That's all we're doing. >> So, the reason why you might find a data set which has large numerical differences between two conditions and yet no statistically significant difference between the two conditions. This is where I'm going to get into trouble. So, the acceptable answer to this question, I'll give you the acceptance first and then I'll give you my kind of alternative answer. The acceptable answer to this question is because basically your confidence intervals are too large for your things that you're comparing. So imagine that you're comparing two different strength training workouts and you've got recovery kind of in the post-workout period and you look at them and it looks like one recovers really quickly and the other one doesn't. But there is no statistical difference between the two.

109:36 There is no significant difference between the two. Now the problem is this resonates with this terminology issue because if somebody who doesn't know what the word significance means in this context looks at those explanations and sees two massive differences sees two recovery profiles they're massively different and then they read the paragraph and it says they're not statistically sign sorry they read the paragraph and it says they're not significantly different statistically speaking then and go what is significant then because those differences are huge numeric direct.

110:08 The issue is that you've got really wide confidence intervals around each of your kind of point measurements that you've taken in the post-workout period such that they overlap and you've ended up in a scenario where essentially you're not sure whether the difference between those two kind of strength training workouts is real or whether it's just a fluke, whether it's just random. >> Actually, no. Now, that's the that's the safe answer. That's the kind of like people aren't well people disagree with everything I say, but like statist statistics people aren't going to object too heavily to that as a as a kind of a generic kind of basic explanation to lay people. Where I'm going to get into trouble is that the problem is with statistics is that the way that you construct your test affects whether you're likely to find a difference or not.

111:02 >> Mhm. So let me give a very very very simple example of this. If I construct a study with only two comparisons. So imagine that imagine that I'm looking only at strength at 24 hours post-workout. So I've got a pre strength training strength measurement and I've got a measurement at 24 hours. And I've got two different workouts. I've got, you know, heavy loads and light loads or whatever. Okay. So, I've got my essentially I've got two strength training sorry, I've got two strength tests for each of those two workouts.

111:44 So, that g me four total data points. Yep. If I do a statistical comparison on those four data points, I am highly likely to find a difference between the two workouts. Okay. If on the other hand I now collect a whole bunch of extra measurement points and I collect you know 0 248 72 96 whatever and I put all of those into a statistical comparison and I've stuffed a whole load of extra data in now the likelihood of me finding a difference has just gone down quite >> that's just how stats works. So the more comparisons you do the less likely it is you're going to find a difference >> and that's what's going to get me into trouble because I'm not supposed to say that. Mhm.

112:24 >> That's just how the calculations work. >> But if you then zero in on each of those individual >> No, because when you do a post hawk calculation, you then have to take into account that you're doing a post talk calculation and your statistical significance thresholds come down. >> Right? >> Okay. Right. That's the issue. So basically, this is the tradeoff that you have when you when you include a control group. If you're doing a study and you're like, oh, I'm not happy this study didn't have a control group. Like, and sometimes you really need a control group. Like, if you've got familiarizations going on, you need a control group. I mean, like, I talked about motor imagery literature recently. If you don't have a control group in a motor imagery study, you are stuffed because all you're going to measure is a familiarization effect and you're not going to know it. So, sometimes you really, really want a control group. But if you put a control group into your calculations and all they do is they rock up to the the laboratory and they sit and watch everybody else do the do the workout and do the strength tests and all they do is that you know participate in the strength test. They don't do any of the workout or anything like that. and literally you just get a whole bunch of measurements and they're just kind of sitting around baseline for the whole kind of protocol and you're feeling very proud of yourself for having had a control group and gone to all that extra effort. you've just destroyed your statistical significance capability because you've now got extra comparisons. And this is the thing that people don't talk about and they're like, you know, but it happens.

113:50 >> Interesting. Yeah. Okay. >> So, I try to discreetly take that into consideration >> without saying that I'm doing that. I get into all kinds of trouble cuz technically you're not supposed to do that. but yeah, so if I see that someone's done an absolute metric ton of comparisons, then I'll work through the statistical calculation that they've done, at least as far as I can establish, and I'll see whether I think that the way that they've done it has prejudiced the experiment in the way that I think they've made it harder to see comparisons happen. You know, I mean, generally speaking, it does.

114:28 >> Presumably, if a researcher is doing a study, they probably want to find an effect most of the time. So why would there be an incentive to not to add extra comparisons and dil? >> Well, as I said, in the control group scenario where you add a control group, you're doing that because you want to make the study more robust. And in some scenarios, >> honestly, the study just doesn't work if you don't have a control group. Like, >> yeah, sure, >> motor imagery stuff, it it literally the the familiarization effect makes it so you just can't see what you're doing.

114:59 but you know many other scenar because you're using untrained subjects. I mean that's the that's the problem. >> You untrained subjects and you have motor control kind of tests. They're going to get better every single time they come back to the test. That's just how human beings work. >> So you need a control group. You can't get away with that. Without >> sometimes you don't need these. >> Exactly. And it's knowing when you can do without it. That's But that's study design. I'm not >> okay. So sometimes it may just be like mistakes that people are making >> as opposed to intentionally adding extra outcomes, >> but you know, you might re I mean like I'm I go crazy for studies that have like loads of time course data. I mean like they're my favorite studies where you see changing over time, but generally speaking in those time course studies you get very very bad statistical significance.

115:48 >> Yeah. Okay. >> Because you got so many measurements. I mean like I was literally looking at a study today and I think they got a measurement every couple of hours for like well apart from the sleep period they got a measurement every couple of hours for like best part of a day and a half >> and they'd missed like a 12-hour chunk for for sleep and then they come back and start taking measurements again. You know I go crazy for those kind of studies. I think they're fantastic but getting statistical significance in those studies is a nightmare.

116:15 >> So would you just run your own stats on that then? Like what is that? you can't because you don't have a data set. But you you try and at least you try and at least be aware that that can happen. >> You mentioned before like there could be a significant a non-significant result, but there could be numerically a large change. >> Yeah. And as I say, mostly that's because you've got a really wide confidence interval because basically you got a huge error rate in your or inaccuracy in your in your results that you're taking >> and so you don't know that it's down to chance. Now, that feels to me like at best what one would say is we don't know if it's down to chance, but it feels like that's then being interpreted as there's therefore no effect.

116:59 >> Yeah, you can't do that. I mean, that's that's the issue. If you've got a really big numerical difference between two, this is the issue with the muscle swelling data. Like, I've kind of thrown all the muscle swelling data into a a chart to show people that, you know, numerically the numbers are huge. like you know kind of anywhere up to 10% even after kind of 48 72 hours post-workout in strength trained people and none of those lines drop below the zero mark. If it was random you'd expect them to be distributed randomly around zero more or less >> just kind of as a sense check and they're not. They're all flying up in the air between two and 10%. You know, now what happens is people go, "Oh, we've got no statistical significance in the in the differences.

117:45 You know, we got no statistically elevated levels of you know, muscle size in the post-workout period. Therefore, it's not happening." I'm like, just look at the numbers and throw them all in a spreadsheet together and you'll see that they're all way above zero, >> way above baseline for the entire post-workout period. You you you're basically just kind of covering your eyes and refusing to look at it. If you think that that's not a signal like you know but again like the way the studies are designed prejudices them towards not generally detecting differences. Errors are you know pretty high and they're taking a ton of measurement >> and like it varies depending on the way you do your statistical calculation but generally speaking it does bias it in that direction. Okay. So, part two coming up on stats.

118:36 >> Not doing a stats. I'm not a stats. I mean, like I can run stats. I've run stats for published papers. I mean, like I can do it. you know, and I know kind of basically how it works. But, you know, I'm not a stats guy really. I'm a physiology guy. And if something doesn't make physiological sense, then I don't care how good your stats are. You know, it's wrong. It's yeah, it's definitely talking to you over the last couple years has definitely opened my eyes up to the the stats side of things when it comes to this and not enough that I am super motivated to learn statistics.

119:16 enough that I've got some statistics books sitting on my shelf and I've opened them. But it is definitely, you know, unfortunately, like you made the comment before, like if you actually want to get into this stuff, it's a 10- year journey to learn all of this. And unfortunately, I think stats is part of that journey to actually really meaningfully engage with this. Hey, >> I would say that the easiest way to learn stats to a point where you know enough to kind of navigate a paper in which stats have been used is to find one of those textbooks that lets you do worked examples yourself. So find find and like even better if you I mean if you can throw it I mean like I I'm comfortable running stuff in R. So I will throw stuff into R and I'll do statistical calculations in R and I'm like that's fun. I can get it to read directly from Excel and so I can have a data set in Excel. I can get R to read it and I can just kind of manipulate the data directly in the in the computer program. you know but like having the ability to just take a data set that you've got and play with it yourself and do calculations on it I think is the most informative frictionless way to learn stats because it is meaningful to you because I think the problem with trying to learn stats when you're just learning like set of rules like if then do that if this then do that >> it doesn't teach you anything in a way that is meaningful to you.

120:58 >> Whereas if you have a data set that you've collected or if you you know kind of pull it off the internet or whatever and you've got this set of numbers and you actually want to try and answer questions about it then suddenly it becomes useful and then you can actually then use the stats in a way that's telling you something. I think that's suddenly when it for me doing that heaven knows how long ago was the point when I actually started to engage with stats in a meaningful way.

121:26 >> I suspect lots of researchers do the same thing. They get to a point where they've got a data set and they're like okay how can I learn about this data set and it's like well you know you're going to need to do stats. I'm like okay well you know I'm going to need to learn these specific areas of stats in order to interrogate my own data set. And I think that's the motivation that probably a lot of people lack if they are not doing this through the research.

121:50 >> Yeah. I've I I have in fact started a course on statistics motivated by our conversations and something I find interesting is these people who are statistics people seem to love statistics and like you made a comment ago that you have you've had fun doing running these statistics yourself and it is it is an observation I've made where yeah people are into stats they they really do end up enjoying them like it is >> it's fun it's fun especially as I say especially if you and get to a point where you've got a computer program that can directly read your data set and you can, you know, kind of manipulate stuff and think about it and decide what tests are appropriate and what are not and you know that kind of thing. It's fun.

122:36 but yeah, at the end of the day, going back to what I said at the beginning, it's like if you forget what your data set is and your data set reflects a physiological reality >> and if you ignore the sources of variance and I know what they say they say, oh no, we tested physers, my physiological prior tell me you can't combine these studies because one's measuring that and another one's measuring something different then I don't care whether you've detected no sources of variance I don't care I think like you can't do it it's qualitative if not quantitatively >> okay if you geek out on hypertrophy research then hopefully you have found this an a helpful episode a little bit different to our normal episodes but hey you know we're over a year in and we're talking about studies all the time and it's was probably time to help people learn how to actually make sense of what they're eating. So hopefully this was a helpful episode to you. If not and you just like lifting heavy weights in the gym, then don't worry, we'll be back to our normal schedule soon enough. What are you shaking your head for me?

123:52 >> I can't imagine anybody's got to this point if they're not at least somewhat interested. >> If they're not interested in Yeah. There's a they used to say or they they say that in Plato's academy he had a sign up on his door and I think it said something like the man without mathematics cannot enter something like that. And I think that's that's a sign above the door of the start of this podcast. If you're not interested in knowing the technicalities of this, you haven't made it in. You haven't made it this far. So if you are here, thank you for making it 2 hours in. We appreciate you and we'll be back next week with a new topic.